Over the past 50 years, the cost of health care in the United States has dramatically increased, equaling $2.9 trillion and equating to 17.4% of the Gross Domestic Product (GDP) in 2013. Imaging costs have also significantly increased over the past several years, a cumulative 70% between 2000 and 2010. A large portion of this imaging cost increase was due to an increase in advanced imaging such as CT and MR, which more than doubled from 2000 to 2010. This increase has occurred despite the lack of studies documenting improved patient outcomes and value from these imaging studies. This rise in imaging, particularly that of CT, has also led to significant increases in radiation exposure to patients and potential increased incidence of radiation induced malignancies. To ensure the appropriate and cost effective use of imaging ? to make sure that imaging is performed the right way, at the right time, for the right patient's ? there is a critical need to perform comparative effectiveness research (CER). One reason for the lack of research in imaging CER is that training in CER, and in the use of biomedical big data that is inherent to CER in imaging, is only available to a very small subset of medical imagers, and is available only at significant cost. Creating effective training in CER and big data analytics for the broader community of medical imagers is beyond the capabilities of individual imaging training programs. In order to address the unique and important need for CER and big data training for imagers, we propose a collaborative and broadly accessible tiered program in CER training and the use of biomedical big data. Tier 1 is a set of 8 lectures on the fundamentals of CER and big data to be available online and also presented at the American Institute for Radiologic Pathology, an intense, 1 month training course attended by 90-95% of all radiology residents in the United States and Canada. Tier 2 is an advanced CER and biomedical big data training program including the potential for continued mentorship in CER and the use of big data, directed toward and available to the medical imaging community. To ensure that this program has the greatest impact, we will use a hybrid educational structure with both online and in-person interactive sessions. Our proposal is unique in that it already has the support of the major national imaging organizations (including the American College of Radiology, American Roentgen Ray Society, Radiological Society of North America, Society of Chairs of Academic Radiology Departments, Association of Program Directors in Radiology, Association of University Radiologists, American Society of Neuroradiology, Society of Skeletal Radiology) as well as members of the imaging industry (including Philips Healthcare and Siemens Medical Solutions).